Safety of pimecrolimus in infants

The favorable safety profile of pimecrolimus in infants has been demonstrated in the above described vehicle-controlled trials (Table 1 ) (30, 46-48, 50, 51). Four of these trials were also evaluated in a previously pooled safety analysis (66). In this safety analysis, a total of 495 pimecrolimus-treated infants were compared with a total of 193 infants who were treated with vehicle (66). Although results of this analysis may have been confounded by larger patient numbers in the pimecrolimus treatment group, rates of adverse events (AEs) were similar between pimecrolimus and vehicle, except for teething which was more commonly observed in pimecrolimus-treated infants (66). All other AEs were disorders typically encountered during childhood, such as nasopharyngitis, pyrexia, upper respiratory tract infections, and bronchitis (66). Burning and erythema were the most frequently reported application site reactions, occurring in less than 1% of infants treated with pimecrolimus (66). Consistent with the low systemic exposure of pimecrolimus in infants (Table 2 ), the pooled safety analysis confirmed that use of pimecrolimus is not associated with immunosuppression (66). Accordingly, the incidence of non-skin infections was comparable between pimecrolimus and vehicle treatment groups (relative risk: 1.015; 95% confidence interval [CI] 0.88-1.18). The incidence rates of bacterial, fungal, parasitic, and viral skin infections were also not altered by pimecrolimus treatment (66). To further evaluate the impact of topical pimecrolimus on the developing immune system, seropositivity rates upon vaccination were assessed in 91 infants, who had applied pimecrolimus for a median number of 377.5 days during 2 years of follow-up (67). Results demonstrated that topical application of pimecrolimus in infants did not interfere with tetanus, diphtheria, measles, and rubella vaccination response (67).
Despite numerous studies demonstrating that pimecrolimus was well-tolerated in infants without evidence of immunosuppression, initial safety and tolerability concerns remained due to the lack of long-term safety data. Additional safety data were requested mainly to determine the impact of pimecrolimus on the developing immune system, and to counteract the link between pimecrolimus use and (skin) malignancies. Since the implementation of the black-box warning, the 5-year PETITE study, several epidemiological studies, post-marketing surveillance, and the ongoing Pediatric Eczema Elective Registry (PEER), all aimed to resolve these safety concerns (58, 68-73). The PETITE study reassured that long-term use of topical pimecrolimus in infants was not related to an increased incidence of 18 AEs, which were of primary clinical interest because of safety signals reported in previous trials (58). Over the 5-year study period, there were no differences in growth rates between both treatments and not more than 1% of patients discontinued their treatment. Similarly to the results from the vehicle-controlled trials, use of pimecrolimus did not affect vaccination response of the infants. Tetanus, hepatitis B, measles, varicella, and Hemophilus influenzae type B antibody titers post vaccination were all considered normal and comparable between infants who had used pimecrolimus or TCS for the past 5 years. The PETITE study provided additional evidence that use of pimecrolimus has no impact on the developing immune system by demonstrating similar humoral and cellular immune responses with pimecrolimus and TCS (58). Furthermore, ex vivo studies showed that levels of CD3+ T-cells over time were comparable not only between pimecrolimus and TCS, but also with historical controls. No malignancies were observed in the pimecrolimus treatment group, and there were no cases of cutaneous T-cell lymphoma or any other skin malignancy in any of the almost 2,500 enrolled infants, out of which more than 1,000 had used pimecrolimus on the long-term (58).
Additionally, five epidemiological studies were conducted to evaluate the potential malignancy risk that is associated with the use of TCIs in general, including pimecrolimus (Table 3 ) (68-70, 72, 74). However, when interpreting results from these studies one should consider that patients with AD are generally more prone to develop cancer compared to patients without AD. This was demonstrated by a study performed in the United Kingdom (UK), comparing more than 60,000 patients with AD across all ages with more than 4 million individuals without AD. After adjusting for age and sex, patients with AD had an increased risk to develop cancer in general (75). Specifically, AD was found to be associated with increased incidences of lymphoma, melanoma, and nonmelanoma skin cancer (NMSC) (75). Even when keeping this in mind, none of the five epidemiological studies, involving more than 6.5 million patients with AD, showed an increased risk of lymphoma, including T-cell cutaneous lymphoma, in patients with AD who had been treated with pimecrolimus (Table 3 ). Consistent with these results, a meta-analysis published by Legendre et al. (2015) failed to find a correlation between pimecrolimus and risk of lymphoma when used for the treatment of AD (Figure 1 ) (76).The same was true for tacrolimus and low-potency TCS, whereas high-potency TCS were associated with a small but significant increase in lymphoma risk (Figure 1 ) (76). However, there was a high level of heterogeneity between the included studies and the results for high-potency TCS may have been biased by disease severity, as high-potency TCS are mainly used for the treatment of severe cases, which in itself is a risk factor for lymphoma (68). Similarly, there was no association between topical use of pimecrolimus and the incidence of NMSC based on a questionnaire that was sent out to 5,000 adult patients with AD out of which 1,000 had NMSC (77). Interestingly, the latter case-control study reported less cases of NMSC in patients treated with TCIs, especially with pimecrolimus (77). Because epidemiological data are retrospectively collected and studies generally had a limited follow-up time, the PEER was initiated in 2004 to prospectively gather post-marketing data in children with AD between the ages of 2-17 years (71). To be included in the registry, children had to be treated for at least 6 weeks with pimecrolimus prior to enrolment into the registry. The primary aim of this nationwide, observational registry in the US and Canada was to evaluate the possible link between topical use of pimecrolimus and malignancy over a follow-up period of 10 years (71). At the time of the first interim results, 7,457 children with AD were enrolled in the PEER for more than 25,000 person-year of follow-up. Over the 10-year study period, five malignancies had been reported (2 leukemia, 1 osteosarcoma, and 2 lymphomas) (71). These malignancies were not considered drug-related and the standardized incidence ratio for all malignancies as compared with the Surveillance, Epidemiology and End Results (SEER) data was 1.2 (95% CI 0.5-2.8), indicating that children exposed to pimecrolimus for the treatment of AD are not at risk to develop cancer (71). As of March 2020, only five additional cases of malignancies were reported in the PEER, evaluating data derived from 8,015 patients and including 45,960 person-year of follow-up (Meda data on file). Importantly, neither the interim results nor the latest follow-up data, reported any cases of skin cancer. Taken together, there is currently no clinical reason to expect malignancies when using topical pimecrolimus for the treatment of AD.